8001.qxd 5/6/04 2:32 PM Page 1237 Academy Report Position Paper Diagnosis of Periodontal Diseases* This position paper on the diagnosis of periodontal diseases was prepared by the Research, Science and Ther- apy Committee of the American Academy of Periodontology. It is intended for the information of the dental profession and other interested parties. The purpose of the paper is to provide the reader with a general overview of the important issues related to the diagnosis of periodontal diseases. It is not intended as a com- prehensive review of the subject. J Periodontol 2003;74:1237-1247. laque-induced periodontal diseases are mixed associated with connective tissue attachment loss also infections associated with relatively specific lead to the resorption of coronal portions of tooth- Pgroups of indigenous oral bacteria.1-6 Suscepti- supporting alveolar bone.16 bility to these diseases is highly variable and depends This simple separation of plaque-induced peri- on host responses to periodontal pathogens.7-11 odontal diseases into two categories is not as clear- Although bacteria cause plaque-induced inflammatory cut as it first appears. For example, if sites that have periodontal diseases, progression and clinical char- been successfully treated for periodontitis develop acteristics of these diseases are influenced by both some gingival inflammation at a later date, do those acquired and genetic factors that can modify sus- sites have recurrent periodontitis or gingivitis super- ceptibility to infection.12-15 imposed on a reduced but stable periodontium? There are currently no data to definitively answer TRADITIONAL APPROACH TO DIAGNOSIS this question. However, since not all sites with gin- Despite our increased understanding of the etiology givitis necessarily develop loss of attachment and and pathogenesis of periodontal infections, the diag- bone,17 it is reasonable to assume that gingivitis can nosis and classification of these diseases is still based occur on a reduced periodontium in which ongoing almost entirely on traditional clinical assessments.16,17 attachment loss is not occurring. A similar problem To arrive at a periodontal diagnosis, the dentist must exists when the term “periodontitis” is assigned to rely upon such factors as: 1) presence or absence of sites with attachment loss and periodontal pockets clinical signs of inflammation (e.g., bleeding upon in which ongoing periodontal destruction is not probing); 2) probing depths; 3) extent and pattern of occurring. loss of clinical attachment and bone; 4) patient’s Demonstration of the progression of periodontitis medical and dental histories; and 5) presence or requires documentation of additional attachment loss absence of miscellaneous signs and symptoms, occurring between at least two time points. Since this including pain, ulceration, and amount of observable is not always possible, especially when a patient is plaque and calculus.18-20 examined for the first time, most clinicians assign the Plaque-induced periodontal diseases have tradi- diagnosis of “periodontitis” to inflamed sites that also tionally been divided into two general categories based have loss of attachment and bone. This is a prudent on whether attachment loss has occurred: gingivitis practice since such sites may be either currently and periodontitis. Gingivitis is the presence of gingi- progressing or are at an increased risk for further val inflammation without loss of connective tissue periodontal destruction. Therefore, demonstration of attachment.16 Periodontitis can be defined as the pres- progressive attachment loss is not generally consid- ence of gingival inflammation at sites where there ered to be a requirement for using “periodontitis” as has been a pathological detachment of collagen fibers a diagnostic label. from cementum and the junctional epithelium has At the 1999 International Workshop for Classification migrated apically. In addition, inflammatory events of Periodontal Diseases and Conditions, a reclas- sification of the different forms of plaque-induced periodontal diseases was developed.21 This revised * This paper was developed under the direction of the Research, Science and Therapy Committee and approved by the Board of Trustees of the classification includes seven general types of plaque- American Academy of Periodontology in May 2003. induced periodontal diseases: 1) gingivitis, 2) chronic J Periodontol • August 2003 1237 8001.qxd 5/6/04 2:32 PM Page 1238 Academy Report periodontitis, 3) aggressive periodontitis, 4) periodon- pathogens in the subgingival flora might be useful in titis as a manifestation of systemic diseases, 5) necro- identifying a microbial target of periodontal therapy, tizing periodontal diseases, 6) abscesses of the but it does not provide information that is used in periodontium, and 7) periodontitis associated with determining a periodontal diagnosis. endodontic lesions.21 The major departures from the previous classification system are: 1) the term “chronic SCIENTIFIC EVALUATION OF DIAGNOSTIC periodontitis” has replaced “adult periodontitis” and TESTS 2) the term “aggressive periodontitis” has replaced Statistical validation of a potentially useful diagnostic “early-onset periodontitis.” In the new classification test routinely involves use of a two-by-two decision system, depending on a variety of circumstances, all matrix as shown in Figure 1. From such tables, the forms of periodontitis can progress rapidly or slowly validity of a diagnostic or prognostic test can be esti- and can be non-responsive to therapy. It was also mated.22 A diagnostic device or test is intended to acknowledged that gingivitis can develop on a reduced detect the presence of a specified disease. Data col- but stable periodontium.21 lection to evaluate a diagnostic test frequently employs The above classification should not be confused a cross-sectional sampling scheme, and the validity of with case types previously suggested by the American the test can be estimated by calculating its sensitivity Academy of Periodontology for purposes of third- and specificity. These can only be determined in a party insurance payments. The current case types cross-sectional study if the true disease status of the for periodontal diseases include: gingivitis (Case patient can be established from a single examination. Type I), mild periodontitis (Case Type II), moderate This is the case for the presence or absence of peri- periodontitis (Case Type III), advanced periodontitis odontitis. The sensitivity of a diagnostic test refers to (Case Type IV), and refractory periodontitis (Case the probability of the test being positive when the dis- Type V). ease is truly present. A perfect test would be able to detect the disease in all cases without registering a DIAGNOSTIC INFORMATION false negative. The sensitivity of such a perfect test Periodontal diagnoses are determined by analyzing would be 1.00. The specificity of a diagnostic test refers the information collected during a periodontal exam- to the probability of the test being negative when the ination. A decision is then made regarding the dis- disease is not present. A perfect test would be able to ease category that is most closely associated with correctly identify all instances in which the disease was the patient’s clinical status. The information routinely absent without registering a false positive. The speci- collected during a periodontal examination includes ficity of such a perfect test would be 1.00. However, in demographic data (e.g., age, gender, etc.), medical medicine and dentistry, perfect diagnostic tests do not history, history of previous and current periodontal exist. Therefore, a test’s sensitivity and specificity will problems, periodontal probe measurements (i.e., always be less than 1.00. It is reasonable to expect probing depths, clinical attachment loss, etc.), radio- that a clinically useful diagnostic test for periodontal dis- graphic findings, and miscellaneous clinical fea- eases should have high values for both sensitivity and tures or observations (e.g., gingival inflammation, specificity. There are, however, no preset upper and plaque/calculus, mobility, occlusal problems). In some lower limits of sensitivity and specificity values that situations, supplemental qualitative or quantitative determine if a diagnostic test is clinically useful. assessments of the gingival crevicular fluid (GCF) and Furthermore, since sensitivity and specificity values subgingival microflora are performed. In addition, a are calculated in diseased or healthy populations, genetic test for susceptibility to chronic periodonti- respectively, these values may be higher than calcula- tis has become commercially available.16 tions performed in a mixed population. In contrast, pre- It should be emphasized that, at the present time, dictive values are calculated in a mixed population of supplemental information on GCF components, the diseased and healthy patients. subgingival microflora, and genetic susceptibility are The positive predictive value of a test refers to the not commonly used by practitioners in arriving at a probability that the disease is present when the test is diagnosis since the diagnostic utility of this inform- positive. The negative predictive value refers to the ation has not been validated. Indeed, genetic testing probability that the disease is absent when the test is is primarily intended to assist in risk assessment and negative. However, predictive values are influenced by should not be considered a diagnostic test.